*Article* **Longitudinal Trends of Participation in Relation to Mental Health in Children with and without Physical Di**ffi**culties**

**Ai-Wen Hwang 1,2 , Chia-Hsieh Chang <sup>3</sup> , Mats Granlund 4, Christine Imms 5, Chia-Ling Chen 1,2 and Lin-Ju Kang 1,2,\***


Received: 5 October 2020; Accepted: 15 November 2020; Published: 18 November 2020

**Abstract:** Children with physical disabilities (PD) are known to have participation restrictions when in inclusive settings alongside typically developing (TD) children. The restrictions in participation over time may affect their mental health status. This study aimed to investigate the longitudinal relationship between independence in activities (capability) and frequency of attendance in activities, in relation to perceived mental health status in children with and without PD. The participants were a convenience sample of parents of 77 school children with PD and 94 TD children who completed four assessments with a one-year interval between each assessment. Parents of these children were interviewed with the Functioning Scale of the Disability Evaluation System—Child version (FUNDES-Child). Three dimensions of mental health problems—loneliness, acting upset, and acting nervous—were rated by parents with the Child Health Questionnaire (CHQ). Linear trend was tested by repeated-measure ANOVA. The results revealed different longitudinal patterns of independence and frequency of attendance over time for children with PD and TD. Frequency of attending activities may be more important than independence in performing activities for experiencing fewer mental health problems. The findings highlight the need for supporting children's actual attendance in daily activities which may benefit their later mental health.

**Keywords:** participation; longitudinal study; physical disabilities; inclusion; mental health

#### **1. Introduction**

Participation, referring to functioning in everyday life beyond the health condition or disability-related diagnosis, is aligned with inclusive education in the Sustainable Development Goals (SDGs) as part of a United Nations Resolution that are intended to be achieved by 2030. SDGoal 4 states that inclusive and equitable quality education and promotion of lifelong learning opportunities in the home, school, and community "for all" must be ensured. Thus, children with disabilities have the same right to education and learning as other children. This SDG goal supports that all children should

be educated within their best-fit environment, providing learning opportunities within participatory learning processes. Therefore, investigating whether the need for positive experiences and learning are met by the unique environmental requirements of children with disabilities will provide critical information for building a society for all.

Physical disability (PD) is one of the categories of disabilities defined in the overall objectives of the United Nations Convention on the Rights of Persons with Disabilities (UNCRPD). Typically, we address the need for safety and equality of school for children with physical difficulty with their peers in an inclusive physical environment [1]. However, the children's mental health, especially in an inclusive setting, is usually not explicitly supported by the surrounding adults and peers. Mental health has been defined as "a state of wellbeing in which every individual realizes his or her own potential, can cope with the normal stresses of life, can work productively and fruitfully and is able to make a contribution to his or her community." [2] School-aged children with physical disabilities [3] and young adults [4] are more likely to develop mental health problems, such as depression and anxiety, than their peers without disabilities.

Research reported increased vulnerability to poor mental health when adolescents make the transition to young adulthood [5]. The mental health of children with a physical disability aged 6–12 years is less well known. In Taiwan, caregivers and professionals focus largely on interventions to improve physical functioning, but mental health is seldom a focus of interest. However, the family costs associated with a mental disorder or mental illness are likely to be higher than those associated with chronic physical disorders [6].

Participation in everyday life activities can be seen as containing two dimensions; physical/virtual attendance and involvement [7]. The life situations in which these dimensions are experienced change over time which influences patterns of attendance and involvement. Long-term outcomes of attendance and involvement may with time affect mental health for children. Mental health may on the other hand affect the probability of adapting to environmental changes following from transition to new life roles. The two dimensions of participation have a bi-directional relation with internal factors within the child as well as external factors in the environment [7]. Internal factors concern activity competence, sense of self, and preferences, while external factors concern physical and social factors in the environment [7]. In earlier research and pediatric rehabilitation intervention, internal factors such as body functions and activity performance have been the focus with the implicit rational that by improving child skills the child will participate more. Thus, activity competence in terms of capability to perform activities in everyday life activities rather than participation in everyday life has been the focus of both assessment and intervention [8–10]. However, the evidence that intervention focusing solely on improving skills leads to enhanced participation is weak [9]. The relationship between activity competence, defined as capability, and the two dimensions of participation needs to be further investigated.

The physical and social activity competence of an individual can be investigated on a continuum from capacity (the ability to perform an activity under ideal circumstances) to capability (the ability to perform an activity in natural environments). In measures of activity competence, e.g., Pediatric Evaluation of Disability Inventory (PEDI) [11] or Child and Adolescent Scale of Participation (CASP) [12], activity competence is operationalized as independence, that is, the level of support needed to perform an activity. The Functioning Scale of the Disability Evaluation System—Child version (FUNDES-Child) [13,14] is a measure containing the further development of CASP to include a measure of frequency of attendance in activities, in addition to the measure of independence in performing an activity (capability). Thus, FUNDES-Child allows us to investigate the relationship between capability and the attendance dimension of participation.

We know that a low frequency of participation in physical activity can lead to a decrease in activity competence defined both as capacity and capability. The International Classification of Functioning, Disability and Health (ICF) framework has been applied in several longitudinal studies that indicate a bi-directional relationship between participation and body function (mental or physical) [15]. For children with severe physical impairments, the longitudinal prediction of participation by body function is stronger than for children with less severe physical impairments [16]. How mental health problems are related to capability as well as participation has been infrequently investigated. A cross-sectional study reported that participation in physical activities can attenuate the odds of depression in children with cerebral palsy (CP) (the Odds Ratio = 1.9; 95%; the Confidence Interval = 0.7–5.3) [3]. Another cross-sectional study reported a bidirectional relationship between mental health problems and participation for children with and without physical disabilities aged 6–14 years [17]. Studies are lacking about how capability and participation can predict or influence later mental health.

It is likely that environmental factors moderate the relationship between capability and participation and mental health, respectively. Barriers in the environment may result in children with disabilities attending activities less frequently than same-aged peers, although they actually have the capability to perform the activity. Kang et al. found that barriers experienced in social support, such as attitudes from family and community, influenced participation more than the physical design of the school for children with physical disabilities [18]. Based on the reported difference between capability and frequency of attending an activity, Hwang et al. proposed that a measure of the gap between independence and frequency of attendance would reflect the closeness of fit between the environment and the person in relation to children with physical impairments [16]. A small gap would indicate a good fit. In Hwang et al.'s study, capability was defined as independence in performing an activity, and frequency of attendance was an operationalization of the attendance dimension of participation [19,20]. Hwang et al. explored the gap between independence and frequency using a nationwide cross-sectional survey with FUNDES-Child [16]. The data showed that the independence–frequency gap of children with cerebral palsy becomes wider with age and that the gap increased more for children with mild compared to severe impairments. The gap may reflect environmental and personal factors that influence individualized service plans or rehabilitation goals aimed at increasing the children's attendance at activities even if they do not have the capability to perform the activities independently.

Studies are needed to reveal the longitudinal influence of participation outcomes and its impacts on other outcomes. Imms and Adair (2016) in a longitudinal study investigated participation in activities outside the school for 93 children with CP for 9 years. Regarding attendance, the diversity of the activities the children attended, as well as the frequency with which the children attended the activities, decreased over time for recreational, active physical, and self-improvement activities; while attendance in social activities increased over time [21]. Anaby et al. (2019), in an intervention study aimed at increasing participation in community activities by adapting the environment, reported that increased self-rated perception of activity performance was related to increased motor capacity (a measure of activity competence) [22].

The purpose of this study was to investigate the longitudinal relationship between independence (capability) and frequency of attendance in relation to perceived mental health status in children with and without physical disabilities. Three specific aims were addressed to reveal the interactions between capability and attendance over time, and how these interactions relate to mental health status. First, the trajectories of independence, frequency of attendance, and the independence–frequency of attendance gap across four years were analyzed for children with and without physical disabilities. Second, the trajectories of independence, frequency of attendance, and the independence–frequency of attendance gap over time were compared in accordance with children's mental health status. Third, the relationships between independence and frequency of attendance across the four years and mental health problems in the last year were examined.

#### **2. Materials and Methods**

#### *2.1. Design*

A four-year longitudinal descriptive study design was used. We analyzed data from children whose families completed surveys at four time points at one-year intervals. Trained interviewers visited each family to collect data.

#### *2.2. Participants*

The proxy–child dyads were recruited from elementary schools in the northern, middle, and southern parts of Taiwan. The inclusion criteria for children with physical disabilities were (1) children from the first to fifth grade; (2) children with the following primary diagnoses or conditions: Amputation, cerebral palsy, cerebral vascular accident/stroke (vascular brain disorders), congenital anomalies, hydrocephalus, juvenile arthritis, nonprogressive muscular disorders, neuropathy, orthopedic conditions (e.g., scoliosis), spinal cord injury, spina bifida, and traumatic brain injury [23], or those who had movement impairments [24] or neuromuscular disabilities [25]; and (3) that parents provided consent. The inclusion criteria for typically developing children were: (1) Children from the first to fifth grade; (2) children without medical diagnosis relating to developmental disabilities; and (3) that parents provided consent. The ethical approval (no. 100-4201A3) was obtained from the Institutional Review Board in the Chang Gung Memorial Hospital in Taiwan. All the participants provided the signed informed consent. The numbers of participants who completed the interviews from the first to the fourth time points were 119, 98, 97, and 94 for TD children, and 93, 78, 78, and 77 for children with PD, respectively (see Table 1 for demographic data). The attrition rates between the first and fourth time points were 21% for TD children and 17% for children with PD.



*Int. J. Environ. Res. Public Health* **2020**, *17*, 8551




interviewers with the Gross Motor Function Classification System (GMFCS), Manual Ability Classification System (MACS), and the Communication Function Classification System (CFCS) at the first time point test.

#### *2.3. Measure*

#### 2.3.1. Functioning Scale of the Disability Evaluation System—Child Version (FUNDES-Child)

The FUNDES-Child utilizes a proxy format in which parents or other caregivers answer questions about their child's activities in the previous 6 months. The FUNDES-Child was translated and modified from the Child and Family Follow-up Survey (CFFS) [12]. The cross-cultural adaption and validation of FUNDES-Child has been reported elsewhere [13,14,18]. The FUNDES-Child contains four parts: Part I: Physical and emotional health (information on health and the way of moving and communication); Part II: Participation (derived from the Child and Adolescent Scale of Participation); Part III: Body function impairment (derived from the Child and Adolescent Factors Inventory); and Part IV: Environmental factors (derived from the Child and Adolescent Scale of Environment). In this study, we only focused on Parts I and II. General mental health status was measured by one question in the FUNDES-Child Part I, which was: "*In general, how would you describe your child's emotional health and well-being (i.e., the way he or she feels about himself or herself and his or her life)?*" The response was rated as 0 (poor), 1 (fair), 2 (good), 3 (very good), and 4 (excellent).

Participation was assessed using the FUNDES-Child Part II that contains 20 items of children's daily participation in 4 settings: Home, neighborhood/community, school, and home/community living. The scale contains two dimensions: Independence and frequency of attendance [13]. Independence was defined as the chi1d's current level of capability to perform the activity compared to other children of his or her age in the same community. For each item, independence was rated as 0 (independent), 1 (with supervision/mild assistance), 2 (with moderate assistance), 3 (with full assistance). Frequency of attendance was rated with reference to age as 0 (the same or more than expected for age), 1 (somewhat less than expected for age), 2 (much less than expected for age), and 3 (never does). The score was designed to match the coding of the ICF qualifiers, with higher scores indicating more limitations or restrictions in capability and performance. In the FUNDES-Child Part II (participation), therefore, a higher score for independence and frequency of attendance indicates a lower level of independence and a low frequency of attending in the activity. A response of "not applicable" (a child of the same age and in the same community would not be expected to do that activity) was allowed for both dimensions. For example, the item "using transportation to get around in the community" could be rated as not applicable if the child did not need to utilize the transportation system. All items were rated under the condition that children used assistive devices as usual.

As each item in the FUNDES-Child Part II (participation) was on the same ordinal scale with the same anchor points at the extreme end (0–3 points), the two dimensions were comparable based on age-expected independence and frequency of attending. Items rated as "not applicable" were omitted in the scoring [13]. The mean scores for each of the 4 settings of FUNDES-Child Part II (participation) are thus the sum of the scores of all "applicable" items divided by the number of applicable items and then converted to a 0–100 scale for the two dimensions. The trained interviewers could, therefore, interpret the scores within the same directional framework (higher scores represented greater participation restriction and more dependence). A score of 0 on either scale could be interpreted as "doing the same as other children the same age". The reliability of the FUNDES-Child Part II (participation) in children with and without physical disability was examined. Test–retest reliability of 86 parent proxies who were interviewed twice within 2 weeks was established for independence (intraclass correlation coefficient [ICC] = 0.955, *p* < 0.001) and frequency of attendance (ICC = 0.796, *p* < 0.001). Interrater reliability of another 77 parent proxy respondents was established for independence (ICC = 0.994, *p* < 0.001) and frequency of attendance (ICC = 0.860, *p* < 0.001).

#### 2.3.2. Child Health Questionnaire (CHQ)

The CHQ is an internationally recognized general health-related quality of life (HRQOL) instrument that has been rigorously translated into more than 78 languages and standardized for use with children aged 5–18 to assess the child's physical, emotional, and social well-being. There are both parent-reported and child self-completed versions of varying lengths. This study applied the parent-reported 28 (PF28) version at the fourth time point of this study. The CHQ covers three items representing mental health problems: "*During the past 4 weeks, how much of the time do you think your child felt lonely?*", "*During the past 4 weeks, how much of the time do you think your child acted nervous?*", "*During the past 4 weeks, how much of the time do you think your child acted bothered or upset?*" Each item is rated with the Likert scale as 1 (all of the time) to 5 (none of the time); thus, a higher score indicates less frequent mental health problems. The score was then transformed to standardized 0 to 100 scores using the algorithm *(raw score* − *1)* × *100*/*4.* The higher standardized score means better mental condition. The whole scale score can be transformed to a Z-score as described in the manual [26].

#### *2.4. Procedure*

Study flyers and research invitations were distributed to schools and hospitals. The teachers and clinicians informed the researchers about the families who were interested in this study, and then the research assistants contacted the families. Following signed informed consent from the children's proxies, the trained testers visited families at home or another place convenient to the family, such as schools or hospitals. The trained interviewers conducted structured interviews with the proxies to collect all data. To reduce participant attrition over time, thank you letters and an invitation for the next year were sent to participants' schools and hospitals to be distributed to families every year around the time of Christmas or Chinese New Year.

#### *2.5. Data Analysis*

The independence–frequency of attendance gap was analyzed by the score of independence minus the score of frequency of attendance. If the independence–frequency of attendance gap was positive (i.e., independence limitation score > frequency of attendance restriction score, where high scores mean more dependent and restricted), it meant that children attended the activity more frequently than what was expected from their level of independence. If the gap was negative (i.e., independence limitation score < frequency of attendance restriction score), it meant that children attended the activity less frequently than what was expected from their level of independence. The trajectories of independence scores, frequency of attendance scores, and the independence–frequency of attendance gap were graphed to provide a global picture of the changes in patterns across four time points. The mean scores for independence and frequency were plotted on dual Y coordinates (from 0 "as expected for age" to 100 "most dependent" or "most restricted", respectively), illustrating any discrepancy between independence and frequency of attendance scores.

Longitudinal statistical analyses were performed with the Statistical Package for Social Science version 21.0 (SPSS, Inc., Chicago, IL, USA). The changes from the first to the fourth year were examined with a two-way repeated measure ANOVA with a group (PD and TD) by time (the first, second, third, and fourth year) interaction. The trend analysis was performed with repeated measures of ANOVA. The significance of the linear trend was tested by ANOVA for 4 time points, and the between-times sum of squares for the effect of the time point was partitioned into a polynomial trend, namely a linear and higher order trend. The polynomial trend component was tested by an F-ratio (the mean square for linear trend/error term). To deal with the variance inequality, Leven's test for homogeneity was conducted before ANOVA. Welch ANOVA and Games–Howell post hoc analyses were performed if the data failed to meet the equal variance assumption with alpha set at 0.05 (2-tailed). The trajectory of independence score, frequency of attendance score, and the independence–frequency of attendance gap were also graphed.

To address the first and second aims, the above analyses were performed for all children with TD and PD and by the five levels of general mental health status (i.e., excellent, very good, good, fair, and poor) in each group. When analyzing each group based on the level of general mental health status, children with TD who were rated as "very good" and "good" and children with PD who were rated as "very good", "good", and "fair" had adequate sample sizes and thus sufficient statistical power for testing the significance in the trend analysis. For other children, only descriptive statistics were presented. To address the third aim, Pearson or Spearman correlations were used for examining the relationships between independence and frequency of attendance at the first to fourth time points and the mental health problems (i.e., loneliness, upset, and nervous) at the final time point. For exploratory purposes, we focused on correlations that reach a significance level of 0.05.

#### **3. Results**

The scores of independence and frequency of attendance measured by the FUNDES-Child Part II (participation) were significantly lower for children with PD than children with TD at each time point (*p* < 0.001). Patterns of change over the four time points showed that the children with PD had increasing scores (i.e., were more dependent and restricted) with age; while children with TD had decreasing scores (i.e., were less dependent and restricted) on the two dimensions (Table 2 and Figure 1). The independence–frequency of attendance gap scores for the TD children were initially negative; they tended to attend activities less frequently although they could perform the activity independently. With time the gap decreased and the change reached significance. For children with PD, the gap was positive, and they tended to be less dependent in the activity, although they attended the activity relatively frequently. With time, the gap increased but did not reach significance (Table 2).

**Figure 1.** Independence and frequency of attendance gap by groups (TD vs. PD) across the four time points. Note: Dark point (•) and black line with 1 SD error bar illustrate the frequency of attendance scores; open circle point (-) and gray line with 1 SD error bar illustrate the independence scores.

For the change patterns in independence and frequency of attendance across levels of general mental status, the children with PD had increasing scores (more limitations and restrictions) with time, while children with TD had decreasing scores (fewer limitations and restrictions; see Table 2 and Figure 2). The gap scores for the TD children were negative. With time, the gap decreased and the change reached significance for children whose mental health status was "very good" and "good". For children with PD, the gap was positive. With time, the gap increased and the change reached significance only for children whose mental health status was "good" (Table 2). The interaction effects of time and group are available in the Supplementary Materials, Table S1.




**Figure 2.** Independence and frequency of attendance gap by groups (TD vs. PD) across four time points by the five mental status of children. Note. TD = typically developing children; PD = physical disability; dark point (•) and black line with 1 SD error bar illustrate the frequency of attendance scores; circle point (-) and gray line with 1 SD error bar illustrate the independence scores; dark star sign (\*) beside Time 4 presents a significant trend for frequency of attendance; gray star sign (\*) beside Time 4 presents a significant trend for independence. Dark star sign (\*) beside the case number presents significant independence–frequency of attendance gap trends in that block.

The correlations between independence scores and frequency of attendance scores across the first to fourth time points and the items of mental health problems at the fourth time point are exhibited in Table 3. Overall, all correlations were in the week-to-moderate range (<±0.4), and all but one was negative. Negative correlation coefficients between the independence and frequency of attendance scores and mental health problems scores indicate that more dependence and higher restrictions in attendance were associated with more mental health problems (i.e., lower scores for loneliness, upset, and nervous). In addition, the correlations between frequency of attendance and mental health problems were in general stronger than those for independence. The independence and frequency of attendance scores were correlated with the score for loneliness only for children with PD, and were also correlated with the score for being nervous only for children with TD. The frequency of attendance, but not independence, was correlated with the score for being upset for both children with TD and PD. When the scores of the three mental health items were aggregated to a Z-score, the frequency of attendance was correlated with the mental health scores at all 4 time points for children with PD, and also at second and third time points for TD children. The correlations between the frequency of attendance and loneliness and being nervous were highlighted by scatter plots of the scores at the fourth time (shown in Figure 3). The plots showed that the TD children who were more restricted in frequency of attendance expressed more feelings of being nervous. The children with PD who were more restricted in frequency of attendance expressed more loneliness.

(**a**) (**b**)

**Figure 3.** The scatter plots of correlations between (**a**) lonely and frequency; (**b**) nervous and frequency at the fourth time point.



\* *p* < 0.05; \*\* *p* < 0.01; T1 to T4 stand for the first to fourth time points.

#### **4. Discussion**

This study is unique for using a longitudinal design to investigate long-term changes in two dimensions, capability and frequency of attendance, and the closeness of fit between these two dimensions. This study described participation trajectories for the same group of children, providing strong evidence about their experiences and opportunities over time. The longitudinal investigation on participation for both children with PD and TD added valuable information to what can be deduced from cross-sectional data [16]. In a previous study, cross-sectional data of cohorts of children with PD showed a fluctuating and/or declining trend in participation attendance with age, especially during the transition from elementary school to junior high school [16]. With a longitudinal design, we were able to trace the adaptive process within the same groups of children across different ages.

TD children on average had a decreasing negative gap between independence and frequency of attendance over time. In other words, with age, expected capacity (independence) and performance (participation) were matched. One explanation for this finding is that TD children experienced both increased capability and a stronger self-selection of what activities to attend frequently with age. In contrast, children with PD had decreasing independence over time and relatively stable frequency

of attendance, suggesting that, despite a widening gap in age-appropriate independence, participation remained possible. Knowledge about typical and atypical trajectories could inform professionals in how to support children's participation as a means of promoting both physical and mental health.

In terms of children's general mental health status rated by parent proxies, a majority of children with TD were rated as good to very good; while a majority of children with PD were rated as fair to very good. This indicates that parents perceived a relatively good state of mental health status of school-age children in Taiwan. We would expect that when children with and without PD are in an inclusive environment, children with PD may need individualized strategies to enhance learning and socialization to a larger extent than TD children. Caregivers and educators may need supports in providing a learning and socially enhancing environment that helps children maintain an adequate level of mental health. A universal design for learning may be needed to meet the diverse participation needs that occur in inclusive education settings, thus supporting non-discriminatory and inclusive education.

The trajectories identified for independence and frequency of attendance over four years were related to proxy-rated mental health status. Children with TD and PD who were less dependent and less restricted in attendance were also reported to have higher levels of mental health status. In particular, the highest dependence and restrictions in attending the activities were reported for children with PD rated as having poor mental health. However, the positive gap (i.e., attending more than the capability score suggested that a child could do independently) for children with PD remained over time, especially for children with good mental health. The positive gap may indicate the importance of support from the environment to enable frequent participation in the activities. For TD children, a gradual narrowing of the negative gap (due to sustained ability and increased frequency of attending an activity) was related to having good or very good mental health. At the last data collection point, TD children's ability and frequency of attendance were matched. These findings suggest that school-age children with PD may have different lived experiences and adaptive processes from their TD peers as they age. For children with PD, continued environmental supports to enable children to attend more than their capability suggests may be an important support for maintaining good mental health status.

Though the children with disabilities have struggled with physical as well as emotional vulnerability, most children in this study were reported to have less frequent mental health problems, also indicating a relatively good state of mental health status. The relationship between frequency of attendance and later mental health problems was stronger than the relationship between independence and later mental health problems. Our results suggest that in children with PD, attending the activities more frequently was associated with less frequent feelings of loneliness, which is a positive outcome for social well-being. This suggests that it is essential to provide supports for children with PD to keep attending activities over time, regardless of whether they have limited ability to perform the activities independently. In children with TD, attending the activities more frequently was associated with less frequent feelings of nervousness. Experiences and competencies gained through participation may facilitate children's confidence and mastery in performing the activities and is thus associated with positive emotional well-being [27].

Findings of this study have implications for environment-based interventions to achieve a match or even a positive gap between independence and frequency of attendance in activities for a child. Environment-based interventions focus on finding solutions built on the child's strengths and capability that help to remove physical, social, and institutional or activity demands barriers to participation [15]. The active facilitation of participation for children with PD by adapting the environment may further facilitate the maintenance of children's mental health. In particular, building a social-friendly environment relies on responsive relationships with care providers and teachers. For rights-based inclusion, the educator has a role to support the child to develop stable friendships by taking advantage of the positive characteristics of each child [28]. This would make inclusive education offer learning opportunities that engage every child, so they learn together and cope with each other.

This study highlighted the importance of exploring longitudinal patterns of capability and participation frequency in relation to general mental health status. There are, however, some limitations pertaining to the measures used in this study. In terms of mental health measures, only one item of general mental health status and three questions about mental health problems were used. Further research may explore a broader set of mental health issues that reflect emotional, psychological, and social well-being. In terms of the measures of participation, children's involvement in the activities was not investigated. It is likely that personal feelings and experiences when actually engaging in the activities are affecting children's mental health and well-being. The relationship between involvement and mental health warrants further investigation.

#### **5. Conclusions**

Children with physical disabilities can, presumably with appropriate supports, sustain a high frequency of attending activities despite difficulties with performing the activities independently. Enriched participation experiences may lead to better mental status of children regardless of disability or not. However, the relationships between frequency of attending, independence, and mental health differed between children with and without PD. Children with TD exhibited fewer mental health problems as rated by proxies, and their negative frequency of attending–independence gap narrowed over time. Children with PD still had a wide positive gap after four years of life experiences. Loneliness was related to less frequent attendance for children with PD, while acting nervously was related to less frequent attendance for children with TD. Interventions for promoting mental health status may be designed based on universal strategies that support participation as well as the characteristics of the individual child.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1660-4601/17/22/8551/s1, Table S1: Main and interaction effects of longitudinal statistical analyses for children with PD and TD children.

**Author Contributions:** Conceptualization, A.-W.H., M.G. and C.I.; data curation, A.-W.H.; formal analysis, A.-W.H., M.G., C.I. and L.-J.K.; funding acquisition, A.-W.H., C.-L.C. and L.-J.K.; investigation, A.-W.H. and L.-J.K.; methodology, A.-W.H., M.G., C.I. and L.-J.K.; resources, C.-H.C., C.I., and C.-L.C.; writing—original draft, A.-W.H.; writing—review and editing, C.-H.C., M.G., C.I., C.-L.C. and L.-J.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Ministry of Science and Technology in Taiwan (MOST 101-2314-B-182-088-, NSC 102-2628-B-182-001-MY3) and Chung Gung Memorial Hospital Medical Research (CMRPD1J0071). The APC was funded by Chung Gung Memorial Hospital (BMRPE09).

**Acknowledgments:** The researchers thank all the families for their time participating in the interviews, and all the research assistants.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**


#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Exploring the Impacts of Environmental Factors on Adolescents' Daily Participation: A Structural Equation Modelling Approach**

**Yael Fogel 1,\* , Naomi Josman <sup>2</sup> and Sara Rosenblum <sup>2</sup>**


**Abstract:** Adolescents with neurodevelopmental difficulties struggle to perform daily activities, reflecting the significant impact of executive functions on their participation. This research examines an integrated conceptual model wherein supportive environmental factors in the community, school and home settings explain the children's participation (involvement and frequency) with their daily activities performance as a mediator. Parents of 81 10- to 14-year-old adolescents with and without executive function deficit profiles completed the Participation and Environment Measure for Children and Youth and the Child Evaluation Checklist. A secondary analysis was conducted to examine the structural equation model using AMOS software. The results demonstrated support for the hypothesised model. Supportive environmental demands in school predicted 32% of home participation, and the adolescents' daily performance reflected that executive functions mediated the relationship between them. Together, these findings highlight the school environment as the primary contributor that affects the children's functioning according to their parents' reports and as a predictor of high participation at home in terms of frequency and involvement. This study has implications for multidisciplinary practitioners working with adolescents in general, and in the school setting specifically, to understand meaningful effects of executive functions on adolescents' daily functioning and to provide accurate assistance and intervention.

**Keywords:** daily activities performance; executive function deficit (EFD); home; school; community; supportive factor; structural equation modelling

#### **1. Introduction**

Participation in daily activities naturally occurs when individuals involve themselves in occupations (daily life activities) that have significance and purpose [1]. Within contemporary theory, participation results from the dynamic transactions between an individual and their environment [2]. The World Health Organization's [3] International Classification of Functioning, Disability and Health: Child and Youth (ICF-CY) version also demonstrated that personal and environmental factors affect interactions among body structure and function, performing daily activities and participating in the community. Adolescents who participate in daily activities form strong bonds with their communities and develop their roles in society, which then helps them prepare for adulthood [4].

Since the ICF-CY was developed, there has been a continued effort to refine the understanding of participation and environmental factors that support or inhibit children both with and without disabilities over time [5]. Maciver et al. [6] reviewed the association among environmental and psychosocial factors with participation in school of children aged 4 to 12 years. Their findings supported the hypothesis that participation outcomes are influenced by known contexts and mechanisms. Specifically, Maciver et al. showed that school routines and structures, objects and spaces and peers and adults are representations of the environment (context). Concerning identified mechanisms, Fogel et al. [7] showed

**Citation:** Fogel, Y.; Josman, N.; Rosenblum, S. Exploring the Impacts of Environmental Factors on Adolescents' Daily Participation: A Structural Equation Modelling Approach. *Int. J. Environ. Res. Public Health* **2021**, *18*, 142. https:// dx.doi.org/10.3390/ijerph18010142

Received: 5 November 2020 Accepted: 23 December 2020 Published: 28 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/).

that children with executive function deficits (EFD) faced more barrier factors in the environment than did their peers and found the activities' social and cognitive demands to be the most challenging.

Executive functions (EFs) are a neuropsychological concept referring to a skillset that composes the cognitive process. This skillset allows people to forsake immediate demands to instead achieve long-term goals and thus to organise their behaviour over time [8]. These EFs influence participation and performance in daily life [9], and the performance of most daily activities requires using different EF components. The literature indicated that EFs might serve as an underlying mechanism in neurodevelopmental disorders such as attention deficit hyperactive disorder (ADHD), specific learning disorder and developmental coordination disorder. The contribution of EFs to adolescents' participation [7], scholastic achievements [10] and daily functioning has been reported [11]. The transition from childhood into adolescence often brings a new set of responsibilities and self-regulatory requirements (e.g., in school and social environments) [12] that necessitate adolescents to rely more on this emerging cognitive control.

Recently, Fogel et al. [13] described adolescents with EFD profiles. These adolescents are characterised as impaired when performing complex daily living activities. They often struggle to achieve everyday life goals as efficiently as their peers without EFD. That is, they require considerably more help from adults, need substantially more time to complete tasks and exhibit behaviours that are far more dangerous [13]. Since adolescents with EFD profiles tend to focus on immediate timeframes, they find planning to be a challenge. They also struggle to shift between activities, prioritise essential tasks, manage their time and meet deadlines [14]. These difficulties hinder their effective participation and performance in everyday life, creating a functioning gap between them and adolescents without EFD [11,13,15].

In the existing literature, discussion of the relationships among performance of daily activities, environmental factors and participation is scarce, and the overall picture including clinical implications—is still unclear. Noreau and Boschen [16] dealt with the complex environment of participation interaction. Their results indicated that despite the environment's obvious theoretical impact on participation, its contribution to restricting or facilitating participation has yet to be demonstrated scientifically. King et al. [17] reported the environment's indirect impact on participation by referring to its direct effects. Specifically, their results showed the adolescent's activity preferences and functional abilities, as well as the family's orientations, to be the most important predictors of participation. Moreover, they indicated the need for a more in-depth look at indirect effects to broaden viewpoints and to consider the roles that other environmental and family factors play in what had been presumed to be causal, developmental sequences.

In contrast, Anaby et al. [18] found that the environment played a mediating role. Their findings explained the participation of young children with or without disabilities across community, school and home settings. Anaby et al. proposed and tested one model for each setting using structural equation modelling (SEM). These models explained 50% to 64% of the variance in both involvement and frequency of participation. According to these results, supports and barriers in the environment significantly mediate between the adolescent's personal factors (e.g., health and functional issues or income) and participation outcomes.

Likewise, most other studies on participation showed that children and adolescents with disabilities participate less in daily activities in terms of level of involvement and frequency in all three settings [19–21] and face more inhibiting environmental factors [7,18]. However, questions about the impact of the child's everyday expression of daily activities performance, and how it relates to participation, are still unanswered and need additional research.

This lack of a documented, compelling association between participation and environmental factors denotes how difficult it is to operationalise these constructs [16]. Therefore, this study examines the extent to which factors that support adolescents' environment also influence their participation. It assumes the mediating factor is adolescents' daily activities

performance. To that end, this research uses the Participation and Environment Measure for Children and Youth (PEM-CY) [20,22], which is a reliable, valid and well-documented tool for assessing both participation and environmental factors. Additionally, the study uses the Child Evaluation Checklist (CHECK) questionnaire, which was also found to be valid and reliable, to examine the daily activities performance that reflects EF in young children [23,24] and adolescents [25]. Combining these two questionnaires (completed by the adolescents' parents) connects the current concept presented by the ICF-CY [3], which views children's and adolescents' functioning holistically. It also reflects previous studies' recommendations to examine the complexity of the relationship between participation and environmental requirements and abilities.

This study assumes that support factors in all three environments (community, school and home) may improve adolescents' daily activities performance and thus affect their participation (involvement and frequency) in the various environments. Figure 1 depicts the proposed theoretical model underlying the direct and indirect factors impacting participation. This conceptual model assumes that if the environment is supportive, then the adolescent's daily activities performance will be better and thus will affect their participation.

**Figure 1.** Conceptual model.

#### **2. Materials and Methods**

#### *2.1. Participants*

This study refers to a secondary analysis using data from a previously published study, which detailed the participant inclusion and exclusion criteria [7,13]. In the current study, the data refer to all participants as one group with no separation between adolescents with and without EFD profiles. Specifically, the participants were 81 early adolescents, 10 to 14 years old (*M* = 12.07 years, *SD* = 1.17). Of them, 57 (70.4%) were boys and 24 (29.6%) were girls. In the original study, 41 participants presented with EFD profiles and 40 with typical development (i.e., without EFD profiles). The EFD profiles were defined using the Behavior Rating Inventory of Executive Function (BRIEF) parent [26] and self-reports [27] and WebNeuro assessments [28]. The parents of all 81 adolescent respondents were invited to participate in the study.

#### *2.2. Procedure*

The University of Haifa Ethics Committee approved this study. Both the parents and the participating adolescents signed informed consent forms. Once accepted into the study, the parents completed a demographic questionnaire. The CHECK provided data regarding the daily activities performance reflecting EF and the PEM-CY as the outcome measure.

We tested two proposed theoretical models. Only one SEM fit the data well and successfully tested both the direct and mediated effects of environmental support factors as an observed (measurable) variable in the community, school and home on the theoretical latent variable, participation. It identified the level of involvement and frequency (10 indicators/items for community, 5 for school, 10 for home) as observed variables, as well as

the theoretical latent variable, daily activities performance (consisting of daily functioning and functioning compared to peers).

#### *2.3. Measurement Instruments*

#### 2.3.1. Demographic Questionnaire

Parents completed the demographic questionnaire, providing data on their education and socioeconomics and on the adolescents' age and gender.

#### 2.3.2. Child Evaluation Checklist

A brief screening instrument used to identify children at risk for under-recognised, invisible neurodevelopmental conditions, the Child Evaluation Checklist (CHECK) [24], emphasises small nuances in the performance features of children's daily activities as related to the children's EFs. The CHECK tool includes two parts. The CHECK-A addresses the current level of daily activities performance, especially frequency. Respondents rate agreement with 30 statements on a Likert scale that ranges from 1 (never) to 4 (always). For example, the statements address whether the adolescents properly estimate the task difficulty and whether they complete tasks they take upon themselves. After exploratory factor analysis, four factors were obtained: organisation (body, essentials and social), self-regulation, performance/expression management and activities of daily living. Cumulatively, these four factors produced a 54.05 variance percentage and α = 94 internal consistency.

The CHECK-B compares the adolescents' general daily function to peers. Using ranks from 1 (low) to 5 (high), parents respond to statements that contain phrases such as, "Compared to other children, my child ... " or "In work habits, my child's overall functioning is . . . ".

We calculated an average score for each part and determined internal consistency (CHECK-A, α = 0.96; CHECK-B, α = 0.94). Construct validity was established and documented in [11].

#### 2.3.3. Participation and Environment Measure for Children and Youth

Although parents completed the primary outcome measure, the Participation and Environment Measure for Children and Youth (PEM-CY) [20,22], herein we present the results in terms of the adolescents as the "participants". Part A of the PEM-CY includes 25 items focusing on participation in a diverse range of activities in community (10 items), school (five items) and home (10 items) settings. For each item, parents report the child's participation through three dimensions: (a) level of involvement on a 5-point scale from 1 (minimally involved) to 5 (very involved); (b) participation frequency on an 8-point scale from 0 (never) to 7 (daily); and (c) parents' desire for a change (e.g., in either involvement or frequency) of their child's participation (yes or no). If parents respond yes to a desire for change, they then select whether they want that change in the child's level of involvement or frequency or a wider variety of activities. However, this study did not include the parents' desire for change.

In this study, participation level of involvement and frequency were calculated as the average of all ratings, except those (either involvement or frequency) for which the parent answered never. The PEM-CY's summary score internal consistency for both participation involvement (α = 0.72–0.83) and frequency (α = 0.59–0.70) was moderate to good. Test–retest reliability for all participation and environment summary scores (interclass correlation (ICC) from 0.58 to 0.95) and across items within the instrument's community, school and home sections (ICC = 0.68–0.96) was also reported as moderate to good [29].

The PEM-CY's Part B asked parents if specific environmental features aided or hindered their children's participation in activities in each (community, school or home) setting. When parents reported a feature as an aid, we coded that item as a support factor. If parents reported the feature made things harder (sometimes or usually), then we coded the item

as a barrier factor. The PEM-CY's summary scores internal consistency for both participation involvement (α = 0.72–0.83) and frequency (α = 0.59–0.70) was moderate to good. Test–retest reliability was reported for all participation and environment summary scores (ICC α = 0.58–0.95) and across items within the instrument's community, school and home sections (α = 0.68–0.96) as moderate to good [22].

#### *2.4. Data Analysis*

Using the bootstrapping method, SEM was conducted to examine the mediation model. The bootstrapping procedure's value lies in its ability to process repeated simulations of subsamples from an original database. With this, we could assess the parameter estimate stability and report their values with increased accuracy. Bootstrapping estimates each resampled dataset's indirect effects and determines a confidence interval for these specific indirect effects [30,31]. We analysed the data using SPSS (version 25) and AMOS software. Indices to evaluate the model included chi-square (acceptable when the value is not significant); comparative fit (CFI); non-normed fit (NNFI; adequate values > 0.90 and excellent fit > 0.95); root-mean-square error of approximation (RMSEA; adequate values < 0.08 and excellent fit < 0.06); and standardised root-mean-square residual (SRMR; <0.08) [32]. Level of significance (*p* value) was 5%.

#### **3. Results**

Table 1 presents descriptive statistics and Pearson correlations among the study variables. Results show that significant correlations were found between the participation variables (involvement and frequency) as measured by the PEM-CY and daily activities performance reflecting EF as measured by the CHECK (*r* = 0.22–0.88; *p* < 0.050 to *p* < 0.001).

The SEM provided excellent goodness of fit indices (χ2(21) = 32.08; *p* > 0.05; NFI = 0.95; CFI = 0.98; RMSEA = 0.08; SRMR = 0.08). As depicted in Figure 2, results of this model showed that higher support of environmental demands at school leads to higher daily activities performance (β = 0.61, *p* < 0.001) and relates positively with home participation (β = 0.53, *p* < 0.05). The indirect effect found between support from the school environment and home participation (β = 0.32, *p* < 0.01) means that the daily activities performance is a mediator between environmental demands at school and home participation. The model explained 58% of home participation.

**Figure 2.** Analysis results of conceptual model mediation. Coefficients in bold are significant at *p* < 0.05. \*\* *p* < 0.01, \*\*\* *p* < 0.001.

Table 2 supports Figure 2. It represents the regression coefficients among all model components to describe the size and direction of the relationship between a predictor and the response variable.





Note. \*\*\* *p* < 0.001.

#### **4. Discussion**

This study examined the effects of supportive environment factors upon participation among adolescents with and without EFD through an SEM approach. Applying SEM allowed us to isolate both the direct and indirect paths by which the environmental (community, school and home) settings affect the adolescents' participation across the three settings. Specifically, our results show that supportive environmental factors in school have indirect effects on home participation, while the adolescents' daily activities performance serves as a mediator of this relationship. That is, no *direct* connection was found between environment and participation; rather, they are connected through the adolescents' daily activities performance. These results may indicate that as long as there is no improvement in the adolescents' daily activities performance following the supports they receive at school, we cannot expect a change in the home environment—not in leisure activities, household chores, school preparation or homework.

#### *4.1. Supportive School Environment Demands*

Role performance in the complex high school environment is critical for academic success; poor role performance in academic and social participation creates high risk for student dropout [33]. The school environment can also often create one of the greatest perceived barriers [34,35]. Unlike several prior studies that described participation barriers without considering the environment's facilitating aspects [16], this study focused on supportive factors. According to the current findings, a supportive school setting can encourage students with and without EFD to more effectively express their daily activities performance and increase participation in the home environment. For instance, Wehlage [36] gathered information from 14 secondary schools that were selected based on their successful dropout prevention programmes. His key findings were relevant to the current study's findings. The results suggested that successful schools create a supportive environment that helps students overcome impediments to membership and engagement. Successful programmes matched students' needs and problems and took advantage of students' interests and strengths. Recently, Mann and Snover [15] argued that to maximise

role performance, environmental influence should be viewed as a means to scaffold and develop EF skills. They mentioned the school environments of interest, including administrative and classroom policies, especially regarding their effects on the interplay between person and role performance (both for students and teachers).

Students and teachers' social interactions, as influenced by educational and social values, create the school climate [37]. Increasingly, research has documented the association of prosocial and academic motivation, conflict resolution, altruistic behaviours and selfesteem with positive school climates. As do teacher–student interactions, the social and educational values that influence children's psychological, social and cognitive development also affect the school climate [38]. Such values include the physical environment [39] and safety [40].

In a previous study, Fogel and colleagues [7] highlighted the environmental factors that, according to parents' reports, best predicted adolescents with EFD. For the school environment, these factors include the activity's cognitive and social demands and staff and teachers' attitudes. Fogel et al. found these factors to substantially aid classification of the study population characteristics (i.e., with or without EFD) and prediction of participants' daily functioning. Since adolescents spend a significant amount of time in school, school activities constitute a significant part of their daily routine both academically and socially.

According to the U.S. Department of Education's [41] Safe and Supportive Schools model, the school climate includes three interrelated domains: (a) the school environment (disciplinary, wellness, physical and academic), (b) student engagement (school participation, respect for diversity and relationships) and (c) safety (substance use/abuse, physical and social-emotional). Bradshaw et al. [37] explored this model. Their findings added to the growing research regarding associations between student outcomes and school climates, indicating that the school climate can significantly predict student achievement.

In 1935, Lewin [42] studied environmental influences on people's (especially children's) behaviour. He suggested that all elements of a child's behaviour, such as the environment where the child lives or how the child plays, influence the child's voluntary behaviour and emotions. Lewin expressed that relationships influence which behaviours a child exhibits, and that those behaviours equate to the child's function and environment.

#### *4.2. Daily Activities Performance Reflecting EF*

Executive dysfunction may be one among many contributors to difficulties adolescents experience [43]. Impaired EFs can lead to compromised self-regulation and decision making, as well as difficulty performing complex or novel tasks. This, in turn, can negatively affect academic performance across the adolescents' life span, leading them to become frustrated when their efforts prove ineffectual and unsuccessful and the outcomes are unsatisfactory (e.g., [44,45]). For example, Mann and Snover [15] measured academic performance to examine how EFs can affect students' role performance and found a significant correlation between poor executive functioning and low academic performance, regardless of setting.

Due to the complexity of recognising such difficulties, children with EFD are most often perceived as having behavioural problems, lazy, lacking motivation, manipulating, "doing it on purpose" and other misleading negative descriptors [9]. Unlike cases of cerebral palsy or intellectual disabilities, for example, the condition of adolescents with EFD is not as clear cut—it may seem invisible. There is a discrepancy between what others can see and what is really happening to these individuals. There are no physical signs, and the adolescents have average or above average intelligence. Nevertheless, the children and their families sense "something different than other children" but do not know what it is or why it is occurring [11]. Unrecognised EFD can compound these effects on daily occupational performance, which then can create secondary issues [46]. Thus, adolescents with EFD must be viewed through the expression of their daily activities performance seen past their externalising behaviours to understand their daily functioning and recognise

their specific needs. Additionally, adaptive programmes and interventions to promote their participation must be created.

#### *4.3. Home Participation*

Adolescence is marked by increased autonomy and access to adult activities and decreased dependence on primary individuals (e.g., parents) and organisational supports [47]. Despite the natural processes occurring in adolescence, this study's results show that the model explained 58% of specific at-home setting participation. Previous studies have identified the environment where the child lives and develops [48] as a critical context in which EFs develop [49], suggesting that individual differences in EFs are also associated with the home environment. Typically, this home environment is measured by the nature, frequency and amount of activities parents create for their children to learn [50]. However, few studies have dealt with the relationship between participation in the home setting and EFs among young children. Korucu et al. [51] investigated potential associations among general parenting practices, EF-related activities in the home and children's EFs beyond the home environment. They discussed the potential importance for pre-schoolers to be exposed in the home environment to EF-specific activities. In 2020, Korucu et al. [52] demonstrated a positive association between more enriching home literacy environments with pre-schoolers' EFs, which then relate to mathematic skills and readiness for general academics.

According to the PEM-CY, this home participation includes leisure and play activities, such as video or computer games, indoor games and play, arts and crafts and other hobbies, listening to music or watching TV, and activities that require social interaction, including getting together with others. Activities such as school preparation (e.g., gathering and packing materials, school bags and lunches or reviewing schedules) and doing homework (e.g., assignments, readings and projects) are also included. This is illustrated through a homework example. The process to finish homework assignments is multifaceted. To successfully complete an assignment, the student must initially record it accurately, bring home the materials needed (e.g., textbooks, handouts), allot after-school time to work on (and ultimately complete) the project, possess the skills needed to finish the work and then bring the finished assignment back to school and turn it in. For assignments that require long-term planning (e.g., long-term projects or preparing for exams), that process becomes even more complex [53]. Such typical assignments can overload the weakened EFs of children with disabilities. To finish a homework assignment, students must (a) keep their attention on the task at hand, (b) ignore distractions, (c) make a plan and set objectives, (d) decide on milestones, such as "where to start" and when to complete, (e) consider details as well as the big picture and (f) organise the relevant materials [54]. Children and adolescents with EFD profiles struggle with those kinds of daily activities, similar to previous findings among students with learning disabilities [55]. For instance, Langberg et al.'s [53] findings suggested that the latter task—organising materials—is critical for students with ADHD in their process to complete homework and thus should be prioritised in interventions.

#### *4.4. Limitations and Future Studies*

Despite its important results, this study has limitations. It included only a small sample in a narrow age range. Larger samples with broader age ranges among adolescents with different disabilities might have expanded the information about the relationship between participation and environment factors across settings. Further, this research did not address parental attitudes towards their children's daily functioning, all factors that may affect functioning perspectives or perspectives other than the parents'. Future studies might incorporate the adolescents' perspectives about their daily activities performance, participation and environmental factors. Future research should analyse the school environment factors for efficient assessment and evaluation processes.

#### **5. Conclusions**

The negative, widespread effects of EFD on occupational performance interfere with adolescents' independence in occupations from self-care routines and social interactions to finishing homework and extend into the classroom. Consistently, adolescents with EF issues have been considered as struggling to start a task, understand what the task requires of them, realise they need, and then ask for, help and recognise when they do not have all the necessary information [9].

This study's findings add to the theoretical and practical evidence of components that can assist and improve participation for adolescents both with and without EFD in general and at home specifically. From the users' viewpoint, supportive school environments may include, for example, physically organising the classroom, providing quiet work areas for children who are distracted by various environmental stimuli, establishing small work groups and dividing tasks into stages with increasing levels of difficulty (to give the children a sense of success and motivation for tasks at higher challenge levels). The emphasis should be on allowing the children to acquire self-management skills in academic and day-to-day tasks (work on problem-solving, planning and, especially, control abilities) and adapting the children's abilities (i.e., allowing the children to recognise their strengths and abilities and understand their difficulties). School procedures can be modified to provide relevant adjustments for each child, to be in constant contact with the children's parents and to envision the children and their needs beyond the school framework.

Improved daily activities performance by adolescents with or without EFD can be possible through involvement in a supportive school environment. Assessing the adolescents' daily activities performance can help determine their level of independence in performing everyday activities. It can educate the entire interdisciplinary team, caregivers and families for optimal intervention, discharge coordination and long-term planning.

**Author Contributions:** Conceptualisation, Y.F., S.R. and N.J.; methods, software and formal analysis, Y.F.; preparing original draft, Y.F.; reviewing and editing, N.J. and S.R.; supervision, N.J. and S.R. All authors have read and agree to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was approved by the Faculty of Social Welfare and Health Sciences, University of Haifa ethics committee (approval numbers 253/13).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical restrictions.

**Acknowledgments:** We thank Reanna Hirsh for her help with data collection and Liron Lamash for her useful advice.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


International Journal of *Environmental Research and Public Health*
